[USRP-users] TX and RX paths sharing a coherent LO

Rickard Radio rickardradio at gmail.com
Thu Oct 9 15:42:47 EDT 2014


Thanks for all informative answers - they clarify this issue very well for me.
It will be more interesting and informative to practically experiment with this, given this knowledge.

Rickard

> On 9 okt 2014, at 20:30, Matt Ettus <matt at ettus.com> wrote:
> 
> 
> Every LO in the system, whether for TX or for RX is generated independently by locking to the same master reference, and so every LO will exhibit mutual phase noise to the others.  There isn't a mechanism to generate a single LO and share it between boards.
> 
> To be clear, the frequencies will all be exactly the same, and the phase will be constant between them.  The only thing you lose is that they have different uncorrelated phase noise between them. 
> 
> Matt
> 
> 
>> On Wed, Oct 8, 2014 at 2:44 PM, Rickard Radio via USRP-users <usrp-users at lists.ettus.com> wrote:
>> OK, interesting. 
>> 
>> Is it the exact same situation with two (or more) different transmitters like TX1-TX2 as you explained here with TX-RX, namely that each path always have physically different LO's but they are generated from the same 10MHz reference? 
>> Or can TX1-TX2 (or RX1-RX2) indeed use physically the same LO (i.e. with same phase noise)?  Or is that impossible too?
>> 
>> That is, is there any difference with synched TX-RX compared to synched TX1-TX2 (or RX1-RX2)?  
>> 
>> The thing is that in my experiment I preferably would like to use physically the same LO for all RX's and TX's (i.e., with exactly the same frequency and jitter etc.) but perhaps I need to resort to synched, but still ever so slightly different, LO's for each path.
>> 
>> Regards,
>> Rickard
>> 
>>> On Wed, Oct 8, 2014 at 7:19 PM, Marcus Müller <usrp-users at lists.ettus.com> wrote:
>>> No, the LOs used for mixing are individually generated from the same 10MHz reference; they thus don't drift away from each other, but they don't share the same jitter etc. However, the Ettus daughterboards usually use rather decent synths, and thus the LO quality should be dominated by the quality of the reference clock -- which again, is quite ok, for most applications, and can be further optimized by GPS-disciplining the reference.
>>> 
>>> Usually, that's how close you can get to shared LO if you want independent TX and RX mixers.
>>> 
>>> Greetings,
>>> Marcus
>>> 
>>> 
>>>> On 08.10.2014 19:04, Rickard Radio via USRP-users wrote:
>>>> Sounds good!
>>>> 
>>>> Does this mean it really shares the same LO, i.e., also with the same and
>>>> uniquely occurring phase noise shared to all paths?! (I want that!)
>>>> 
>>>> Regards,
>>>> Rickard
>>>> 
>>>> 
>>>> 
>>>> On Wed, Oct 8, 2014 at 6:54 PM, Marcus Müller <usrp-users at lists.ettus.com>
>>>> wrote:
>>>> 
>>>>>  Hi Rickard,
>>>>> 
>>>>> all the LOs are synthesized from the same motherboard reference, so Q1 is
>>>>> answered with "you get that for free".
>>>>> Q2 is "yes".
>>>>> Q3 gets the answer "if you want to use definite times when sampling, use
>>>>> the timed commands feature of UHD".
>>>>> :)
>>>>> 
>>>>> Greetings,
>>>>> Marcus
>>>>> 
>>>>> On 08.10.2014 18:48, Rickard Radio via USRP-users wrote:
>>>>> 
>>>>> Dear usrp-users,
>>>>> 
>>>>> I am interested in coherently LO-mixing different TX and RX paths for some
>>>>> experiments.
>>>>> 
>>>>> Q1: Which options exist to use the same(?) coherent(!) LO for different TX
>>>>> and RX paths on the USRPs (N210 and/or B210)?
>>>>> Q2: Related, with the N210s and the MIMO cable, can I obtain and use a
>>>>> coherently shared LO for different TX and RX paths?
>>>>> Q3: Follow up  - do I need something else or in addition ?
>>>>> 
>>>>> Regards,
>>>>> Rickard
>>>>> 
>>>>> 
>>>>> 
>>>>> 
>>>>> ______________________________
>>>>> _________________
>>>>> USRP-users mailing listUSRP-users at lists.ettus.comhttp://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
>>>>> 
>>>>> 
>>>>> 
>>>>> _______________________________________________
>>>>> USRP-users mailing list
>>>>> USRP-users at lists.ettus.com
>>>>> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
>>>>> 
>>>>> 
>>>> 
>>>> 
>>>> _______________________________________________
>>>> USRP-users mailing list
>>>> USRP-users at lists.ettus.com
>>>> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
>>> 
>>> 
>>> _______________________________________________
>>> USRP-users mailing list
>>> USRP-users at lists.ettus.com
>>> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
>>> 
>> 
>> 
>> _______________________________________________
>> USRP-users mailing list
>> USRP-users at lists.ettus.com
>> http://lists.ettus.com/mailman/listinfo/usrp-users_lists.ettus.com
>> 
> 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ettus.com/pipermail/usrp-users_lists.ettus.com/attachments/20141009/2b8d8b05/attachment-0002.html>


More information about the USRP-users mailing list